UCSF PDGM: 2D Slice-Based Multi-Sequence Brain MRI with Tumor Segmentation

Derived from the UCSF Preoperative Diffuse Glioma MRI collection, this dataset provides 2D slice-based multi-sequence brain MRI data. It contains preoperative scans of patients with histopathologically confirmed diffuse gliomas, hosted on The Cancer Imaging Archive. The dataset was uploaded by chehablab and was last updated on 2026-06-17.

Use Cases

• Train tumor segmentation models based on multi-sequence 2D MRI slices.
• Develop algorithms for glioma classification and analysis using preoperative imaging data.
• Benchmark medical image processing methods on expert-corrected tumor segmentation labels.
• Study the relationship between multi-sequence MRI features and histopathological glioma types.

Strengths

• Derived from a large-scale, preoperative brain MRI dataset hosted on The Cancer Imaging Archive (TCIA).
• Includes expert-corrected multicompartment tumor segmentations for model training and validation.
• Provides skull-stripped, co-registered, multi-sequence MRI volumes, suggesting high preprocessing quality.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• Freshness should be verified as the last update timestamp is in the future (2026-06-17).

Provenance

Source

The Cancer Imaging Archive (TCIA), specifically the UCSF Preoperative Diffuse Glioma MRI (UCSF-PDGM) collection.

Collection Method

Derived from original 3D MRI volumes; likely processed into 2D slices.

Freshness

Last updated 2026-06-17 11:10:46